[OAI-implementers] XSD file for qualified DC
herbert van de sompel
herbertv@lanl.gov
Tue, 18 Jun 2002 09:32:54 -0600
--------------F8A0500043BC647AD6F9C99A
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
hi all,
Let me try to summarize:
1. In the context of the OAI-PMH, it would make a lot of sense to treat
citations as a parallel metadata format. The unqualified DC record describes
the "paper", whereas another record (under the same item) describes all the
citations made in the "paper". That is what Carl suggested in his mail. And
that is the approach that Stevan Harnad and I discussed at last year's
OAI-related conference in Geneva. This approach makes sense in that it is
extensible: it allows other stuff related to the "paper" (for instance usage
logs, certification metadata, preservation metadata, etc.) to be treated in yet
other parallel records under the same item.
2. When it comes to choosing a "metadata format" to describe those citations,
looking at OpenURL makes a lot of sense. Not only because it is becoming a
standard, but because its purpose really IS to describe stuff (read "citations"
in this context) by building on a broad range of identifier-namespaces and a
multitude of metadata formats. Moreover, OpenURL allows not only for the
description of a "citation" but (optionally) also of entities that make up the
context in which the "citation" appears. That is very significant when thinking
about the possibility of open linking at the level of OAI service providers.
And it is significant when thinking of using "OpenURL" as a parallel metadata
format, as it allows the citation to remain attached to the thing in which it is
cited.
Last weekend, I have been working on an XML encoding that I hope will find its
way into the OpenURL standard (I X my fingers). The thing being encoded is the
context-object. That is the thing described in the Bison-Fute paper
(http://www.dlib.org/dlib/july01/vandesompel/07vandesompel.html), that contains
descriptions of the "cited thing" (referent in OpenURL terminology), the
referring-entity (the thing in which the citation is made), the referrer (the
system that provides the description of the referent), etc. Several
context-objects (read contextual descriptions of cited things) can be bundled in
a context-container. that context-container might be the parallel metadata
format in which citations could be bundled for OAI-PMH. the context-container
format is fomalized by means of an XML Schema at
http://lib-www.lanl.gov/~herbertv/niso/context-container.xsd (experimental!).
Let's take the following example:
The D-Lib Magazine paper:
(a) Van de Sompel, Herbert and Oren Beit-Arie. 2001. "Open Linking in the
Scholarly Information Environment Using the OpenURL Framework." D-Lib Magazine.
7(3). <http://www.dlib.org/dlib/march01/vandesompel/03vandesompel.html> <doi
10.1045/march2001-vandesompel>
Cites:
(b) Caplan, Priscilla and Arms, William Y. 1999. Reference Linking for Journal
Articles. D-Lib Magazine. 5(7/8).
<http://www.dlib.org/dlib/july99/caplan/07caplan.html> <doi
10.1045/july99-caplan>
(c) Van de Sompel, Herbert and Hochstenbach, Patrick. 1999b. Reference Linking
in a Hybrid Library Environment. Part 2:SFX, a Generic
Linking Solution. D-Lib Magazine. 5(4).
<http://www.dlib.org/dlib/april99/van_de_sompel/04van_de_sompel-pt2.html> <doi
10.1045/april99-van_de_sompel-pt2>
In OpenURL terminology:
- paper (b) is a referent
- paper (c) is a referent
- paper (a) is a referring-entity (it references a and b)
- D-Lib Magazine is the referrer, as in this example it provides the "OpenURL"
Using the context-container encoding, the above citations could be rendered as
follows:
Version 1
======
<?xml version="1.0" encoding="UTF-8" ?>
<context-container xmlns="http://www.niso.org/context-object"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://www.niso.org/context-object
http://lib-www.lanl.gov/~herbertv/niso/context-container.xsd"
timestamp="2002-06-14" version="Z39.nn-xml">
<context-object timestamp="2002-06-14T12:13:00Z">
<referent>
<id type="openurl">doi:10.1045/july99-caplan</id>
<id type="uri">http://www.dlib.org/dlib/july99/caplan/07caplan.html</id>
<metadata>
<ja:jarticle xmlns:ja="http://www.niso.org/OpenURL/jarticle/"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://www.niso.org/OpenURL/jarticle/
http://lib-www.lanl.gov/~herbertv/niso/jarticle.xsd">
<ja:ISSN>10829873</ja:ISSN>
<ja:volume>5</ja:volume>
<ja:issue>7/8</ja:issue>
<ja:aulast>caplan</ja:aulast>
<ja:aufirst>priscilla</ja:aufirst>
<ja:atitle>Reference Linking for Journal Articles</ja:atitle>
</ja:jarticle>
</metadata>
</referent>
<referrer>
<id type="openurl">dbid:dlib.org:D-Lib Magazine</id>
</referrer>
<referring-entity>
<id type="openurl">doi:10.1045/march2001-vandesompel</id>
<id
type="uri">http://www.dlib.org/dlib/march01/vandesompel/03vandesompel.html</id>
</referring-entity>
</context-object>
<context-object timestamp="2002-06-14T12:13:00Z">
<referent>
<id type="openurl">doi:10.1045/april99-van_de_sompel-pt2</id>
<id
type="uri">http://www.dlib.org/dlib/april99/van_de_sompel/04van_de_sompel-pt2.html</id>
<metadata>
<ja:jarticle xmlns:ja="http://www.niso.org/OpenURL/jarticle/"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://www.niso.org/OpenURL/jarticle/
http://lib-www.lanl.gov/~herbertv/niso/jarticle.xsd">
<ja:ISSN>10829873</ja:ISSN>
<ja:volume>5</ja:volume>
<ja:issue>4</ja:issue>
<ja:aulast>van de sompel</ja:aulast>
<ja:aufirst>herbert</ja:aufirst>
<ja:atitle>Reference Linking in a Hybrid Library Environment. Part
2:SFX, a Generic Linking Solution</ja:atitle>
</ja:jarticle>
</metadata>
</referent>
<referrer>
<id type="openurl">dbid:dlib.org:D-Lib Magazine</id>
</referrer>
<referring-entity>
<id type="openurl">doi:10.1045/march2001-vandesompel</id>
<id
type="uri">http://www.dlib.org/dlib/march01/vandesompel/03vandesompel.html</id>
</referring-entity>
</context-object>
</context-container>
The same 2 citations could be rendered in a much shorter way:
Version 2
======
<?xml version="1.0" encoding="UTF-8" ?>
<context-container xmlns="http://www.niso.org/context-object"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://www.niso.org/context-object
http://lib-www.lanl.gov/~herbertv/niso/context-container.xsd"
timestamp="2002-06-14" version="Z39.nn-xml">
<context-object timestamp="2002-06-14T12:13:00Z">
<referent>
<id type="openurl">doi:10.1045/july99-caplan</id>
</referent>
<referring-entity>
<id type="openurl">doi:10.1045/march2001-vandesompel</id>
</referring-entity>
</context-object>
<context-object timestamp="2002-06-14T12:13:00Z">
<referent>
<id type="openurl">doi:10.1045/april99-van_de_sompel-pt2</id>
</referent>
<referring-entity>
<id type="openurl">doi:10.1045/march2001-vandesompel</id>
</referring-entity>
</context-object>
</context-container>
greetings
herbert
Ann Apps wrote:
> Carl, Herbert,
>
> I agree with Carl's proposal here. But I think that any XML schemas
> to cover citations should be based on the work Herbert is doing for
> OpenURL, rather than more of us cooking something up. Otherwise
> we would be in danger of duplicating effort again, and OpenURL is
> heading towards standardisation. Then we could include citation
> information within an openurl namespace in OAI-PMH, so we could
> have something like:
>
> <dcterms:citation>
> <openurl:jtitle>Lecture Notes in Computer Science</openurl:jtitle>
> <openurl:atitle>Dublin Core Metadata for Electronic
> Journals</openurl:atitle>
> <openurl:aulast>Apps</openurl:aulast>
> etc
> <openurl:admin_ver>Z39.XX-NN</openurl:admin_ver>
> </dcterms:citation>
>
> or wrapped by <dcterms:references> for references [and making
> assumptions about dcterms:citation, otherwise wrapped by
> dc:identifier]
>
> Herbert, please correct me if I have got this wrong! I assume we
> could use an OpenURL namspace before it becomes a 'real'
> standard. And that we can do this as part of OpenURL version 1.0
> (because it would not have a resolver).
>
> If we utilise an OpenURL schema, we can immediately do citations
> for more than just journal articles, such as conference papers,
> books and parts, patents, etc.
>
> I think it would still need some agreement with DCMI, that the
> OpenURL schema could be used within the particular DC
> elements, but hopefully this would be acceptable because it
> doesn't depend on a DC structured value. [However I do still
> advocate a DCSV for the literal string citation, being less cryptic
> than OpenURL for human reading, but that is outside the scope of
> OAI.]
>
> Best wishes,
> Ann
>
> Subject: RE: [OAI-implementers] XSD file for qualified DC
> Date sent: Thu, 13 Jun 2002 14:53:48 -0400
> From: "Carl Lagoze" <lagoze@cs.cornell.edu>
> To: <ann.apps@man.ac.uk>,
> "Tim Brody" <tim@tim.brody.btinternet.co.uk>
> Copies to: <oai-implementers@oaisrv.nsdl.cornell.edu>
>
> > Interesting discussions from all. As Herbert points out in another
> > mail, this is an opportunity for some interesting cooperative work. I
> > think we all agree that 'citation-type' metadata is useful. Tim's
> > example at
> > http://citebase.eprints.org/cgi-bin/oai?verb=GetRecord&metadataPrefix=
> > dcq&identifier=oai:arXiv:hep-th/0102001 demonstrates how one might
> > make available for harvesting not only the citation data for a
> > resource but its list of references.
> >
> > My personal perspective is that trying to shoehorn these requirments
> > into the DC element structure is probably the wrong approach. It
> > relies on a notion of DC Structured Values, the status of which is
> > pretty uncertain with in DCMI and for which there is no well-defined
> > data model. It also seems to take the approach of DC as a root format
> > on which to build further expressive structrues, rather than as one
> > metadata format among many. Wouldn't a better approach be to exploit
> > the reality we now have realized in OAI: that is, availability of
> > multiple parallel metadata formats? SHouldn't citation and reference
> > metadata be packaged as a metadata format parallel to DC (qualified or
> > unqualified)? Sure there is some duplication of information with DC
> > (e.g., yes a citation has a title) but, taking the view that the
> > metadata format exposed for harvesting is just a database view, I
> > don't think that this is really a relevant argument.
> >
> > So, a proposal. Let's move this reference/citations metadata
> > requirement out from under the notion of 'application profile' for DC
> > or qualifiers for DC and see if there can be some agreement on a good
> > XML schema (profile) for making this information harvestable via
> > OAI-PMH.
> >
> > Probably if people like Tim and Ann put there heads together on this,
> > possibly with the contributions of Simeon Warner and Thomas Krickel,
> > we could come up with something that could fit into a revised version
> > of the OAI implementation guidelines document.
> >
> > Carl
> >
> > Carl Lagoze
> > Senior Research Associate, Computing and Information Science, Cornell
> > University Director of Technology, National Science Digital Library
> > Open Archives Initiative Executive Upson Hall Ithaca, NY 14853 Phone:
> > 607-255-6046 FAX: 607-255-4428 URL: http://www.cs.cornell.edu/lagoze
> >
>
> --------------------------------------------------------------------------
> Mrs. Ann Apps. Senior Analyst - Research & Development, MIMAS,
> University of Manchester, Oxford Road, Manchester, M13 9PL, UK
> Tel: +44 (0) 161 275 6039 Fax: +44 (0) 0161 275 6040
> Email: ann.apps@man.ac.uk WWW: http://epub.mimas.ac.uk/ann.html
> --------------------------------------------------------------------------
--
Herbert Van de Sompel
digital library research & prototyping
Los Alamos National Laboratory - Research Library
+ 1 (505) 667 1267 / http://lib-www.lanl.gov/~herbertv/
objects on mirror sites are closer than they appear
--------------F8A0500043BC647AD6F9C99A
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
hi all,
<p>Let me try to summarize:
<p>1. In the context of the OAI-PMH, it would make a lot of sense to treat
citations as a parallel metadata format. The unqualified DC record
describes the "paper", whereas another record (under the same item) describes
all the citations made in the "paper". That is what Carl suggested
in his mail. And that is the approach that Stevan Harnad and I discussed
at last year's OAI-related conference in Geneva. This approach makes
sense in that it is extensible: it allows other stuff related to the "paper"
(for instance usage logs, certification metadata, preservation metadata,
etc.) to be treated in yet other parallel records under the same item.
<p>2. When it comes to choosing a "metadata format" to describe those citations,
looking at OpenURL makes a lot of sense. Not only because it is becoming
a standard, but because its purpose really IS to describe stuff (read "citations"
in this context) by building on a broad range of identifier-namespaces
and a multitude of metadata formats. Moreover, OpenURL allows not
only for the description of a "citation" but (optionally) also of entities
that make up the context in which the "citation" appears. That is
very significant when thinking about the possibility of open linking at
the level of OAI service providers. And it is significant when thinking
of using "OpenURL" as a parallel metadata format, as it allows the citation
to remain attached to the thing in which it is cited.
<p>Last weekend, I have been working on an XML encoding that I hope will
find its way into the OpenURL standard (I X my fingers). The thing
being encoded is the context-object. That is the thing described
in the Bison-Fute paper (<A HREF="http://www.dlib.org/dlib/july01/vandesompel/07vandesompel.html">http://www.dlib.org/dlib/july01/vandesompel/07vandesompel.html</A>),
that contains descriptions of the "cited thing" (referent in OpenURL terminology),
the referring-entity (the thing in which the citation is made), the referrer
(the system that provides the description of the referent), etc.
Several context-objects (read contextual descriptions of cited things)
can be bundled in a context-container. that context-container might
be the parallel metadata format in which citations could be bundled for
OAI-PMH. the context-container format is fomalized by means of an
XML Schema at <A HREF="http://lib-www.lanl.gov/~herbertv/niso/context-container.xsd">http://lib-www.lanl.gov/~herbertv/niso/context-container.xsd</A>
(experimental!).
<p>Let's take the following example:
<p>The D-Lib Magazine paper:
<p>(a) Van de Sompel, Herbert and Oren Beit-Arie. 2001. "Open Linking in
the Scholarly Information Environment Using the OpenURL Framework." D-Lib
Magazine. 7(3). <<A HREF="http://www.dlib.org/dlib/march01/vandesompel/03vandesompel.html">http://www.dlib.org/dlib/march01/vandesompel/03vandesompel.html</A>>
<doi 10.1045/march2001-vandesompel>
<p>Cites:
<p>(b) Caplan, Priscilla and Arms, William Y. 1999. Reference Linking for
Journal Articles. D-Lib Magazine. 5(7/8). <<A HREF="http://www.dlib.org/dlib/july99/caplan/07caplan.html">http://www.dlib.org/dlib/july99/caplan/07caplan.html</A>>
<doi 10.1045/july99-caplan>
<p>(c) Van de Sompel, Herbert and Hochstenbach, Patrick. 1999b. Reference
Linking in a Hybrid Library Environment. Part 2:SFX, a Generic
<br> Linking Solution. D-Lib Magazine. 5(4). <<A HREF="http://www.dlib.org/dlib/april99/van_de_sompel/04van_de_sompel-pt2.html">http://www.dlib.org/dlib/april99/van_de_sompel/04van_de_sompel-pt2.html</A>>
<doi 10.1045/april99-van_de_sompel-pt2>
<p>In OpenURL terminology:
<br>- paper (b) is a referent
<br>- paper (c) is a referent
<br>- paper (a) is a referring-entity (it references a and b)
<br>- D-Lib Magazine is the referrer, as in this example it provides the
"OpenURL"
<p>Using the context-container encoding, the above citations could be rendered
as follows:
<p>Version 1
<br>======
<p><?xml version="1.0" encoding="UTF-8" ?>
<br><context-container xmlns="<A HREF="http://www.niso.org/context-object">http://www.niso.org/context-object</A>"
<br>
xmlns:xsi="<A HREF="http://www.w3.org/2001/XMLSchema-instance">http://www.w3.org/2001/XMLSchema-instance</A>"
<br>
xsi:schemaLocation="<A HREF="http://www.niso.org/context-object">http://www.niso.org/context-object</A>
<br>
<A HREF="http://lib-www.lanl.gov/~herbertv/niso/context-container.xsd">http://lib-www.lanl.gov/~herbertv/niso/context-container.xsd</A>"
<br>
timestamp="2002-06-14" version="Z39.nn-xml">
<br><context-object timestamp="2002-06-14T12:13:00Z">
<br> <referent>
<br> <id type="openurl">doi:10.1045/july99-caplan</id>
<br> <id type="uri"><A HREF="http://www.dlib.org/dlib/july99/caplan/07caplan.html">http://www.dlib.org/dlib/july99/caplan/07caplan.html</A></id>
<br> <metadata>
<br> <ja:jarticle xmlns:ja="<A HREF="http://www.niso.org/OpenURL/jarticle/">http://www.niso.org/OpenURL/jarticle/</A>"
<br>
xmlns:xsi="<A HREF="http://www.w3.org/2001/XMLSchema-instance">http://www.w3.org/2001/XMLSchema-instance</A>"
<br>
xsi:schemaLocation="<A HREF="http://www.niso.org/OpenURL/jarticle/">http://www.niso.org/OpenURL/jarticle/</A>
<br>
<A HREF="http://lib-www.lanl.gov/~herbertv/niso/jarticle.xsd">http://lib-www.lanl.gov/~herbertv/niso/jarticle.xsd</A>">
<br> <ja:ISSN>10829873</ja:ISSN>
<br> <ja:volume>5</ja:volume>
<br> <ja:issue>7/8</ja:issue>
<br> <ja:aulast>caplan</ja:aulast>
<br> <ja:aufirst>priscilla</ja:aufirst>
<br> <ja:atitle>Reference
Linking for Journal Articles</ja:atitle>
<br> </ja:jarticle>
<br> </metadata>
<br> </referent>
<br> <referrer>
<br> <id type="openurl">dbid:dlib.org:D-Lib Magazine</id>
<br> </referrer>
<br> <referring-entity>
<br> <id type="openurl">doi:10.1045/march2001-vandesompel</id>
<br> <id type="uri"><A HREF="http://www.dlib.org/dlib/march01/vandesompel/03vandesompel.html">http://www.dlib.org/dlib/march01/vandesompel/03vandesompel.html</A></id>
<br> </referring-entity>
<br></context-object>
<br><context-object timestamp="2002-06-14T12:13:00Z">
<br> <referent>
<br> <id type="openurl">doi:10.1045/april99-van_de_sompel-pt2</id>
<br> <id type="uri"><A HREF="http://www.dlib.org/dlib/april99/van_de_sompel/04van_de_sompel-pt2.html">http://www.dlib.org/dlib/april99/van_de_sompel/04van_de_sompel-pt2.html</A></id>
<br> <metadata>
<br> <ja:jarticle xmlns:ja="<A HREF="http://www.niso.org/OpenURL/jarticle/">http://www.niso.org/OpenURL/jarticle/</A>"
<br>
xmlns:xsi="<A HREF="http://www.w3.org/2001/XMLSchema-instance">http://www.w3.org/2001/XMLSchema-instance</A>"
<br>
xsi:schemaLocation="<A HREF="http://www.niso.org/OpenURL/jarticle/">http://www.niso.org/OpenURL/jarticle/</A>
<br>
<A HREF="http://lib-www.lanl.gov/~herbertv/niso/jarticle.xsd">http://lib-www.lanl.gov/~herbertv/niso/jarticle.xsd</A>">
<br> <ja:ISSN>10829873</ja:ISSN>
<br> <ja:volume>5</ja:volume>
<br> <ja:issue>4</ja:issue>
<br> <ja:aulast>van de sompel</ja:aulast>
<br> <ja:aufirst>herbert</ja:aufirst>
<br> <ja:atitle>Reference
Linking in a Hybrid Library Environment. Part 2:SFX, a Generic Linking
Solution</ja:atitle>
<br> </ja:jarticle>
<br> </metadata>
<br> </referent>
<br> <referrer>
<br> <id type="openurl">dbid:dlib.org:D-Lib Magazine</id>
<br> </referrer>
<br> <referring-entity>
<br> <id type="openurl">doi:10.1045/march2001-vandesompel</id>
<br> <id type="uri"><A HREF="http://www.dlib.org/dlib/march01/vandesompel/03vandesompel.html">http://www.dlib.org/dlib/march01/vandesompel/03vandesompel.html</A></id>
<br> </referring-entity>
<br></context-object>
<br></context-container>
<p>The same 2 citations could be rendered in a much shorter way:
<p>Version 2
<br>======
<p><?xml version="1.0" encoding="UTF-8" ?>
<br><context-container xmlns="<A HREF="http://www.niso.org/context-object">http://www.niso.org/context-object</A>"
<br>
xmlns:xsi="<A HREF="http://www.w3.org/2001/XMLSchema-instance">http://www.w3.org/2001/XMLSchema-instance</A>"
<br>
xsi:schemaLocation="<A HREF="http://www.niso.org/context-object">http://www.niso.org/context-object</A>
<br>
<A HREF="http://lib-www.lanl.gov/~herbertv/niso/context-container.xsd">http://lib-www.lanl.gov/~herbertv/niso/context-container.xsd</A>"
<br>
timestamp="2002-06-14" version="Z39.nn-xml">
<br><context-object timestamp="2002-06-14T12:13:00Z">
<br> <referent>
<br> <id type="openurl">doi:10.1045/july99-caplan</id>
<br> </referent>
<br> <referring-entity>
<br> <id type="openurl">doi:10.1045/march2001-vandesompel</id>
<br> </referring-entity>
<br></context-object>
<br><context-object timestamp="2002-06-14T12:13:00Z">
<br> <referent>
<br> <id type="openurl">doi:10.1045/april99-van_de_sompel-pt2</id>
<br> </referent>
<br> <referring-entity>
<br> <id type="openurl">doi:10.1045/march2001-vandesompel</id>
<br> </referring-entity>
<br></context-object>
<br></context-container>
<p>greetings
<p>herbert
<p>Ann Apps wrote:
<blockquote TYPE=CITE>Carl, Herbert,
<p>I agree with Carl's proposal here. But I think that any XML schemas
<br>to cover citations should be based on the work Herbert is doing for
<br>OpenURL, rather than more of us cooking something up. Otherwise
<br>we would be in danger of duplicating effort again, and OpenURL is
<br>heading towards standardisation. Then we could include citation
<br>information within an openurl namespace in OAI-PMH, so we could
<br>have something like:
<p><dcterms:citation>
<br> <openurl:jtitle>Lecture Notes in Computer Science</openurl:jtitle>
<br> <openurl:atitle>Dublin Core Metadata for Electronic
<br>Journals</openurl:atitle>
<br> <openurl:aulast>Apps</openurl:aulast>
<br> etc
<br> <openurl:admin_ver>Z39.XX-NN</openurl:admin_ver>
<br></dcterms:citation>
<p>or wrapped by <dcterms:references> for references [and making
<br>assumptions about dcterms:citation, otherwise wrapped by
<br>dc:identifier]
<p>Herbert, please correct me if I have got this wrong! I assume we
<br>could use an OpenURL namspace before it becomes a 'real'
<br>standard. And that we can do this as part of OpenURL version 1.0
<br>(because it would not have a resolver).
<p>If we utilise an OpenURL schema, we can immediately do citations
<br>for more than just journal articles, such as conference papers,
<br>books and parts, patents, etc.
<p>I think it would still need some agreement with DCMI, that the
<br>OpenURL schema could be used within the particular DC
<br>elements, but hopefully this would be acceptable because it
<br>doesn't depend on a DC structured value. [However I do still
<br>advocate a DCSV for the literal string citation, being less cryptic
<br>than OpenURL for human reading, but that is outside the scope of
<br>OAI.]
<p>Best wishes,
<br> Ann
<p>Subject:
RE: [OAI-implementers] XSD file for qualified DC
<br>Date sent:
Thu, 13 Jun 2002 14:53:48 -0400
<br>From:
"Carl Lagoze" <lagoze@cs.cornell.edu>
<br>To:
<ann.apps@man.ac.uk>,
<br> "Tim Brody" <tim@tim.brody.btinternet.co.uk>
<br>Copies to:
<oai-implementers@oaisrv.nsdl.cornell.edu>
<p>> Interesting discussions from all. As Herbert points out in another
<br>> mail, this is an opportunity for some interesting cooperative work.
I
<br>> think we all agree that 'citation-type' metadata is useful.
Tim's
<br>> example at
<br>> <a href="http://citebase.eprints.org/cgi-bin/oai?verb=GetRecord&metadataPrefix=">http://citebase.eprints.org/cgi-bin/oai?verb=GetRecord&metadataPrefix=</a>
<br>> dcq&identifier=oai:arXiv:hep-th/0102001 demonstrates how one
might
<br>> make available for harvesting not only the citation data for a
<br>> resource but its list of references.
<br>>
<br>> My personal perspective is that trying to shoehorn these requirments
<br>> into the DC element structure is probably the wrong approach.
It
<br>> relies on a notion of DC Structured Values, the status of which is
<br>> pretty uncertain with in DCMI and for which there is no well-defined
<br>> data model. It also seems to take the approach of DC as a root format
<br>> on which to build further expressive structrues, rather than as one
<br>> metadata format among many. Wouldn't a better approach be to exploit
<br>> the reality we now have realized in OAI: that is, availability of
<br>> multiple parallel metadata formats? SHouldn't citation and
reference
<br>> metadata be packaged as a metadata format parallel to DC (qualified
or
<br>> unqualified)? Sure there is some duplication of information
with DC
<br>> (e.g., yes a citation has a title) but, taking the view that the
<br>> metadata format exposed for harvesting is just a database view, I
<br>> don't think that this is really a relevant argument.
<br>>
<br>> So, a proposal. Let's move this reference/citations metadata
<br>> requirement out from under the notion of 'application profile' for
DC
<br>> or qualifiers for DC and see if there can be some agreement on a
good
<br>> XML schema (profile) for making this information harvestable via
<br>> OAI-PMH.
<br>>
<br>> Probably if people like Tim and Ann put there heads together on this,
<br>> possibly with the contributions of Simeon Warner and Thomas Krickel,
<br>> we could come up with something that could fit into a revised version
<br>> of the OAI implementation guidelines document.
<br>>
<br>> Carl
<br>>
<br>> Carl Lagoze
<br>> Senior Research Associate, Computing and Information Science, Cornell
<br>> University Director of Technology, National Science Digital Library
<br>> Open Archives Initiative Executive Upson Hall Ithaca, NY 14853 Phone:
<br>> 607-255-6046 FAX: 607-255-4428 URL: <a href="http://www.cs.cornell.edu/lagoze">http://www.cs.cornell.edu/lagoze</a>
<br>>
<p>--------------------------------------------------------------------------
<br>Mrs. Ann Apps. Senior Analyst - Research & Development, MIMAS,
<br> University of Manchester, Oxford Road, Manchester,
M13 9PL, UK
<br>Tel: +44 (0) 161 275 6039 Fax: +44 (0) 0161 275 6040
<br>Email: ann.apps@man.ac.uk WWW: <a href="http://epub.mimas.ac.uk/ann.html">http://epub.mimas.ac.uk/ann.html</a>
<br>--------------------------------------------------------------------------</blockquote>
<p>--
<br>Herbert Van de Sompel
<br>digital library research & prototyping
<br>Los Alamos National Laboratory - Research Library
<br>+ 1 (505) 667 1267 / <A HREF="http://lib-www.lanl.gov/~herbertv/">http://lib-www.lanl.gov/~herbertv/</A>
<p>objects on mirror sites are closer than they appear
<br> </html>
--------------F8A0500043BC647AD6F9C99A--